1 ID# Exam 1 PS 306, Fall 2004 As always, the Skidmore Honor Code is in effect. Keep your eyes focused on your own exam. At the end of the exam, you will be asked to attest, in writing, to adhering to the Honor Code. Read each question carefully and answer each question completely. I try to construct questions such that the point value of the question reflects the time to answer the question. Thus, you should expect to work for about 15 minutes on a 15-point question. If you spend too long working on any one question, you wouldn’t have sufficient time for other questions. If you spend too little time working on a question, you may well write an answer that lacks detail. So please use your time wisely. Good Luck! 1. Define what we mean by the term operational definition and then cite a specific example of an operational definition from both Milgram's shock study and Harlow's infant monkey study. [5 pts]2 2. A study was conducted to compare the effectiveness of different rewards that might be used to teach retarded children how words are spelled. Forty retarded children, chronological ages 5 to 7, were randomly assigned to four groups, with 10 children in each group. Each child was shown five common objects and five cards, each showing the printed name of one of the objects. The child’s task was to match each object correctly with its name card. Whenever a correct match was made, the experimenter rewarded the child. The type of reward determined the four levels of the factor: Candy Tokens that could be exchanged for candy Tokens that could be exchanged for attention (playing games, reading to the child, etc.) Verbal praise The experimenter recorded the number of trials required before a child could correctly match all five pairs of words and pictures. Thus, smaller numbers indicate faster learning. Analyze and interpret the output below as completely as you can. [15 pts] 10 12.400 2.836 .89710 11.300 2.263 .71610 14.500 3.171 1.00310 13.700 2.003 .633Count Mean Std. Dev. Std. Err.CandyToken/AttentToken/CandyVerbal praiseMeans Table for Column 2Effect: Reward3 3. First of all, define external validity and contrast it with internal validity (be very clear!!). What would you describe as the major point that Mook was making in his article on external invalidity? Mook argues that external validity may not be all that important when we are considering: a. whether something can happen, rather than whether it typically does happen; b. a prediction that something should happen in the lab, based on a theory; c. showing that a phenomenon is so powerful that it can happen even under unnatural conditions. Using specific studies as examples, briefly describe the evidence from the article that illustrates two of the above conditions under which external invalidity may not be all that problematic. [10 pts]4 4. What is the difference between Type I and Type II Errors? What is power? With which type of error is power most closely associated? Why? [5 pts] 5. What is a non-manipulated characteristic of a participant? (Examples help.) Briefly describe a (hypothetical) study using such a non-manipulated characteristic as the “independent variable.” Then, tell me explicitly why you would not be comfortable making a causal claim about any results that emerged from such a study. [5 pts]5 6. Moor and McCabe (1989) looked at the relationship between expenditures on alcohol and tobacco in data from the Family Expenditure Survey of the British Department of Employment. The data are as seen below: Region Alcohol Tobacco North 6.47 4.03 Yorkshire 6.13 3.76 Northeast 6.19 3.77 East Midlands 4.89 3.34 West Midlands 5.63 3.47 East Anglia 4.52 2.92 Southeast 5.89 3.20 Wales 5.27 3.53 Scotland 6.08 4.51 Below you will find StatView analysis of this data set. Interpret the results as completely as you can. If you knew that expenditure on tobacco was 6.0, what would be your best prediction of expenditure on alcohol? If you knew that expenditure on tobacco was 4.02, what would be your best prediction of expenditure on alcohol? As it turns out, there was also data from Northern Ireland. For that area, the expenditure on alcohol was 4.02 and the expenditure on tobacco was 4.56. What would the inclusion of this area have on the overall analysis? (In other words, if you were to add this data point to the scattergram, what would happen to the correlation?) {Howell} [10 pts] 100.784.615.567.446CountNum. MissingRR SquaredAdjusted R SquaredRMS ResidualRegression SummaryAlcohol vs. TobaccoRow exclusion: Untitled Dataset #11 2.543 2.543 12.785 .00728 1.591 .1999 4.135DF Sum of Squares Mean Square F-Value P-ValueRegressionResidualTotalANOVA TableAlcohol vs. TobaccoRow exclusion: Untitled Dataset #12.041 1.001 2.041 2.038 .07591.006 .281 .784 3.576 .0072Coefficient Std. Error Std. Coeff. t-Value P-ValueInterceptTobaccoRegression CoefficientsAlcohol vs. TobaccoRow exclusion: Untitled Dataset #14.254.54.7555.255.55.7566.256.56.75Alcohol2.6 2.8 3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6TobaccoY = 2.041 + 1.006 * X; R^2 = .615Regression PlotRow exclusion: Untitled Dataset #16 7. A habituation task is often used to test the memory of infants. In the habituation procedure, a stimulus is shown for a brief period, and the researcher records how much time the infant spends looking at the stimulus. This same process is repeated again and again. If the infant begins to lose interest in the stimulus (decreases the time looking at it), the researcher can conclude that the infant “remembers” the earlier presentations and is demonstrating habituation to (boredom with?) an old familiar stimulus. Hypothetical data from a habituation experiment were analyzed to produce the analyses seen below. Infants were exposed to stimuli of Low, Medium, and High Complexity. In the course of the experimental session, infants saw each stimulus 6 times for 1 minute each, with the order of presentation varied in a counterbalanced fashion (e.g., LMH, LHM, MLH, etc.). The dependent variable of interest here is the time spent looking at each type of stimulus on the sixth (and last) presentation of each stimulus. Complete the source table and then analyze and interpret the data as completely as you can. [15 pts] 10 .744 .116 .03710 .890 .073 .02310 .966 .048 .015Count Mean Std. Dev. Std. Err.Low ComplexityMedium ComplexityHigh ComplexityMeans Table for ComplexityEffect: Category for Complexity7 8. What is